> Forward Deployed Engineers · Join the Team
Build at the frontier.
> Become one of the people who can deploy frontier AI inside consequential organizations.
We're building a team of exceptional Forward Deployed Engineers who want to work directly on difficult, consequential AI deployments: regulated voice agents, infrastructure for frontier model providers, autonomous operations inside real companies.
The role is unusual because of the combination it demands. Our FDEs can move from an ambiguous business problem to architecture, implementation, and production while working directly with the people who will use what they build.
The combination the role demands
- Deep technical ability
- You build production systems yourself. Backend, data, infrastructure, and the AI layer on top of them.
- AI model knowledge
- You know what current models can and cannot do, how to evaluate them, and when classical ML is the better answer.
- Architecture
- You can design the system, choose the integration points, and defend the trade-offs to a security team.
- Production engineering
- Observability, evals, rollback, on-call. You have run what you built and been paged for it.
- Product judgment
- You can tell which part of a workflow should be automated, which should stay human, and why.
- Customer judgment
- You work directly with executives and end users, and you know when to push back.
- Comfort with ambiguity
- You can start from a business problem with no spec and arrive at a system in production.
- Ownership of outcomes
- The deployment working in production is the deliverable. You own that, not a ticket queue.
What the work looks like
Embed with a customer team as the engineer accountable for an AI initiative reaching production.
Work across architecture, models, integrations, evaluation, infrastructure, and the workflow around the system.
Talk directly to business owners, engineers, security, and leadership. No account managers in between.
Carry what you learn from one deployment into the next, and into the tooling the whole team uses.
> We don't hand you a strategy deck and disappear. We deploy — and we run what we build in production, or train your team to.
The kind of systems you would deploy
> Recent production deployments by our FDEs.